The Authority’s role in making public interventions on the use of statistics
Introduction
The Statistics and Registration Service Act 2007 established the UK Statistics Authority with the statutory objective of “promoting and safeguarding the production and publication of official statistics that serve the public good”.
The Office for Statistics Regulation (OSR), the Authority’s independent regulatory arm, is responsible for undertaking monitoring to identify issues and investigating concerns raised with the Authority. We call this type of work casework. We use our voice to stand up for statistics and to represent the public, reporting publicly where we have concerns and highlighting good practice.
This policy sets out how OSR decides when it will or will not intervene in cases where there are specific concerns about statistics, especially their use in public debate. To find out more about our casework process, how to raise a concern and frequently asked questions, visit the casework pages on the OSR website.
Scope
Our statutory remit covers official statistics produced by the UK governments. Official statistics are statistics produced by Crown bodies (such as UK government departments) or those specified in statutory orders, as defined in the Statistics and Registration Service Act 2007.
We consider the use of statistics and data by organisations who produce official statistics as well as members of governments across the UK. We also expect parliamentarians outside of government in each nation to adhere to the same standards we expect from ministers when quoting statistics in public debate.
We recognise that public trust in the use of data extends beyond official statistics, and the distinction between official statistics and other data may not always be clear. Where we judge that our intervention can support public trust in the use of data in public debate, particularly in high-profile cases, we will intervene on an advisory basis.
Concerns we may consider beyond our usual area of focus include (but are not limited to):
- The use of unpublished management or operational information
- Numbers based on evaluation or modelling exercises
- Commitments to spending or levels of resources allocated to a policy area
- The presentation of statistics in research commissioned by UK governments
- Forecasts and projections such as future outturns of economic statistics or population projections
When we consider cases that are beyond our statutory remit but relate to the presentation of quantitative information in the public domain, any intervention will be made on an advisory basis and will focus on how the use of these numbers complies with the Standards for the Public Use of Statistics, Data and Wider Analysis. We believe it is essential to provide guidance to support public confidence in the appropriate use of data, and to help ensure that public debate is informed by transparent and trustworthy information. See our frequently asked questions for more information on cases beyond our scope.
We reserve the right to not intervene in cases outside of outside our statutory remit, especially where there are other organisations better placed to deal with those issues. In all cases, we will be clear about what we will and will not do, or who we suggest may be better placed to deal with the issue, explaining our reasoning for our judgement.
Interventions
The topics covered by casework can be varied. They could relate to production, communication or use of statistics or compliance with the Code of Practice for Statistics. We are guided by our commitment to statistics that demonstrate trustworthiness, quality and value and will be proportionate in judging when we intervene.
When concerns are raised with us about the production or publication of statistics, we will base our interventions on our Standards for Official Statistics. In some cases, we will not intervene through casework, but the issue raised will prompt us to undertake some other regulatory work, such as a compliance review or an assessment. Our decision to intervene will take into account:
- Whether the issue relates to official statistics (including accredited official statistics or official statistics in development)
- The impact of the issue
- Whether the issue has already been raised directly with the organisation responsible for producing the statistics
When considering issues around the use of statistics, we will base our interventions on our Standards for Public Use of Statistics, Data and Wider Analysis, which are built on the principles of intelligent transparency:
- Equality of access – Public bodies should make statistics, data and wider analysis used in the public domain available to everyone in a timely way so that the public can easily access, scrutinise and verify claims and decisions made based on them.
- Supporting understanding – Public bodies should use statistics, data and wider analysis with integrity and communicate them with clarity and accuracy so that the public can easily understand the basis for claims and decisions made.
- Decision making and leadership – Public bodies should seek and use impartial, expert advice when using statistics, data and wider analysis in the public domain so that the public can have confidence that they have been used appropriately
We recognise that the context is important, and we do not seek to be a mediator in political debate. It is not for us to take a view on decisions made based on statistics. When making judgements relating to the extent to which a claim is misleading, we will use our definition of misleadingness:
“In considering whether documents or statements are liable to mislead, we are concerned when, on a question of significant public interest, statistics are used to communicate a descriptive statement that the wider relevant statistical evidence would not support, despite otherwise being an accurate statement.”
Public reporting
The majority of interventions made by OSR are private. These interventions focus on driving positive change within organisations to improve the statistical system and resolving queries from individual members of the public.
In determining whether to publish the outcome of an intervention, we consider the value in making correspondence public and the likelihood of the public being misled on an issue of significant public interest. When cases are raised on behalf of organisations, we may also publish correspondence for transparency if the organisation has chosen to publicly share the correspondence themselves. We will always maintain the confidentiality of individuals who raise concerns with us, and no identifying details will be published without their permission.
Published or private correspondence will usually come from a member of the Office for Statistics Regulation, usually the Director General for Regulation. When cases are raised by parliamentarians or the issue relates to broader political use, the Authority Chair will respond.
During formal pre-election periods when the publication of new outputs and policies is restricted, OSR can continue intervening publicly under its role of promoting and safeguarding the production and publication of official statistics. However, we will only intervene publicly where a statement is materially wrong or misleading to the point it could significantly impact on the political debates that are presented to voters.
Feedback
All cases are recorded in our issues log – whether internally generated or externally raised, and whether our response is published or not.
The UK Statistics Authority Board keeps its practices under regular review. If you have any feedback or queries, please contact regulation@statistics.gov.uk.
Who responds
Published or private correspondence will usually come from a member of the Office for Statistics Regulation, usually the Director General for Regulation. When the issue relates to broader political use, especially by Ministers and elected representatives, the Authority Chair will respond.
The UK Statistics Authority Board keeps its practices under regular review. If you have any feedback or queries, please contact regulation@statistics.gov.uk.